Chemoenzymatic Hydroxymethylation of Carboxylic Acids by Tandem Stereodivergent Biocatalytic Aldol Reaction and Chemical Decarboxylation

Article abstract of DOI:10.1021/acscatal.9b01646

Chiral 2-substituted 3-hydroxycarboxylic acid derivatives are valuable building blocks for the preparation of naturally occurring and synthetic biologically active molecules. Current methodologies for the preparation of these compounds are still limited f

Full text of DOI:10.1021/acscatal.9b01646

Subscriber access provided by UNIVERSITY OF TOLEDO LIBRARIES
Article
Chemoenzymatic Hydroxymethylation of Carboxylic Acids by Tandem
Stereodivergent Biocatalytic Aldol Reaction and Chemical Decarboxylation
Roser Marín-Valls, Karel Hernández, Michael Bolte, Jesús Joglar, Jordi Bujons, and Pere Clapés
ACS Catal., Just Accepted Manuscript • DOI: 10.1021/acscatal.9b01646 • Publication Date (Web): 08 Jul 2019
Downloaded from pubs.acs.org on July 9, 2019
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 37
ACS Catalysis
1
2
3
4
5
6
7
8
9
Chemoenzymatic Hydroxymethylation of
Carboxylic Acids by Tandem Stereodivergent
Biocatalytic Aldol Reaction and Chemical
Decarboxylation
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Roser Marín-Valls1, Karel Hernández1, Michael Bolte2, Jesús Joglar1, Jordi Bujons1,
Pere Clapés1*
1Instituto de Química Avanzada de Cataluña IQAC-CSIC. Jordi Girona 18–26, 08034
Barcelona, Spain.
2Institut für Anorganische Chemie, J.-W.-Goethe-Universität, D-60438 Frankfurt/Main,
Germany.
*Phone +34934006127, Fax: +34932045904, email: pere.clapes@iqac.csic.es.
ACS Paragon Plus Environment
1

ACS Catalysis
Page 2 of 37
1
2
3
4
5
6
7
8
9
ABSTRACT. Chiral 2-substituted 3-hydroxycarboxylic acid derivatives are valuable
building blocks for the preparation of naturally occurring and synthetic biologically active
molecules. Current methodologies for the preparation of these compounds are still limited
for large-scale production due to the high costs, limited microbial strains, low yields,
difficult downstream processing, and limited range of structures. We report an effective
chemoenzymatic method for the synthesis of enantiomerically pure 2-substituted
3-hydroxycarboxylic esters. The strategy comprises: i) a stereoselective aldol addition of
2-oxoacids to methanal catalyzed by two enantiocomplementary 2-oxoacid aldolases, ii)
oxidative decarboxylation, and iii) esterification. Compounds with S-configuration were
obtained in 69-80% isolated yields (94-99% ee), and the R enantiomers in 57-88% (88-
95% ee), using a substrate concentration range of 0.1-1.0 M. The method developed
offers a versatile alternative route to this important class of chiral building blocks, and
highlights the exciting opportunities available for using natural enzymes with minimal
active site modification.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment
2

Page 3 of 37
ACS Catalysis
1
2
3
4
5
6
7
8
9
KEYWORDS Biocatalysis, 2-Substituted 3-Hydroxycarboxylic acids, 2-Oxoacids, Aldol
Reaction, 2-Oxoacid Aldolases, 2-keto-3-deoxy-L-rhamnonate aldolase (YfaU, EC
4.1.2.53), 3-methyl-2-oxobutanoate hydroxymethyltransferase (KPHMT, EC 2.1.2.11)
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
INTRODUCTION
Enantiomerically pure 2-substituted 3-hydroxycarboxylate esters are an important class
of chiral building blocks widely used in organic synthesis.1-5 For example, (S)- and (R)-
methyl 3-hydroxy-2-methylpropanoate (i.e., Roche esters) are largely employed as
precursors for the preparation of bioactive natural compounds and pharmaceuticals (e.g.,
discodermolide with potent antimitotic activity; epothilone analog ZK-Epo (sagopilone),
the ionophore antibiotic zincophorin, cytotoxic marine natural products auripyrones A and
B, the depsipeptide stevastelin B3 and the antifungal polyketide soraphen A).6-22 Other
chiral 2-substituted 3-hydroxycarboxylate esters can be found in the preparation of
peptide deformylase inhibitors (Figure 1).3,5
ACS Paragon Plus Environment
3

ACS Catalysis
Page 4 of 37
1
2
3
4
5
6
7
8
O
O
CO2Me
S
N
O
OH
OTHP
OH
O
OH
O
CH3
ZK-Epo (Sagopilone)
CH3
9
O
CHO
N
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
R1
HO
N
CO2Me
HO
F
Oxazolidine LBM-415
analog
O
N
O
N
O
H
O
OBn
N
OBn
H
N
HO
OH
N
N
CO2Me
H
O
O
O
O
Macrocyclic, peptidomimetic inhibitor
of peptide deformylase
Figure 1. Examples of 2-substituted 3-hydroxycarboxylate esters as chiral building
blocks for the synthesis of naturally occurring and synthetic bioactive molecules.
A straightforward access to these important compounds would imply the asymmetric
-hydroxymethylation of carboxylate esters using methanal via aldol addition reaction.
Evans chiral 2-oxazolidone auxiliary intermediate is an available stoichiometric non-
catalyzed strategy, but with limited possibility of scale up.5,23,24 Catalytic stereoselective
methodologies include (Figure 2): i) hydroformylation of 1-alkenes,25 with moderate
regioselectivity, ii) Baylis-Hillman reaction between methyl acrylate and methanal,
followed by stereoselective reduction of the resulting methyl 2-(hydroxymethyl)acrylate
using enzymatic or metal catalysis, which is limited to 3-hydroxy-2-methylpropanoate,26-
ACS Paragon Plus Environment
4

Page 5 of 37
ACS Catalysis
1
2
3
4
5
32
iii) organocatalytic asymmetric addition of an aldehyde to methanal furnishing the 2-
6
7
8
9
substituted 3-hydroxyaldehydes and requiring an ensuing chemical transformation (e.g.,
Wittig reaction, oxidation of the aldehyde) due to the delicate nature of these
compounds,2,33,34 and iv) access to (S)-3-hydroxy-2-methylpropionic acid by conversion
of isobutyric acid or directly from glucose utilizing Pseudomonas taiwanensis VLB120
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
B83 T7 cells as catalyst, with limited range to structures with natural constitution.9
A. Hydroformylation of 1-alkenes
TBSO
O
Rh(acac)(CO)2/(S,S)-Ph-BPE
150 psi CO/H2
O
TBSO
H
+
TBSO
H
4:1
96% ee
B. Baylis-Hillman reaction and enzymatic or metal reduction
O
O
OH
O
OH
O
OH
O
H
H
Reduction a) or b)
OMe
OMe
or
OMe
OMe
DABCO
95-99% ee
a) (S) or (R) ene-reductases/NADPH/NAD or b) Rh or Ru/Ligand/H2 (60 bar)
C. Organocatalytic methodology
O
O
OH
O
H
H
O
O
aq
Pinnick Ox.
R2
H
OH
OH
R3
R3
OR4
R1
R1
R2
R1
R2
N
90-99% ee
H
D. Fermentative biosynthesis
Whole cells
OH
O
ee not reported
Glucose
OH
E. This Work: stereodivergent chemoenzymatic synthesis
O
CO2Na
O
OH
OH
O
R1
H
a) KPHMT
b) and c)
a) MBP-YfaU
b) and c)
R2O
OR2
+
O
R1
R1
93-95% ee
H
94-99% ee
ACS Paragon Plus Environment
5

ACS Catalysis
Page 6 of 37
1
2
3
4
5
6
7
8
9
Figure 2. Methodologies for the synthesis of Roche ester derivatives. A) Hydroformylation
of 1-alkenes.25 B) Baylis-Hillman and ene-reductases: Old Yellow Enzyme family (OYE,
EC 1.6.99.1),26-28,31 or rhodium-catalyzed stereoselective hydrogenation of α-
(hydroxymethyl)-acrylate derivatives.29,30 C) Organocatalytic methodology by an ensuing
chemical transformation.2,33,34 D) Fermentative biosynthesis of (S)-3-hydroxyisobutyric
acid from glucose using Pseudomonas taiwanensis VLB120 B83 T7.9 E) This work:
stereodivergent chemoenzymatic synthesis; a) aldol addition; b) decarboxylation, c)
esterification.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Herein, we report a robust and effective three-step chemoenzymatic method for the
synthesis of enantiomerically pure 2-substituted 3-hydroxycarboxylate esters (4)
(Scheme 1). The first step consists of an aldol addition of 2-oxoacids (1) to methanal (2)
catalyzed by two enantiocomplementary 2-oxoacid utilizing aldolases from E. coli:
2-keto-3-deoxy-L-rhamnonate aldolase (YfaU, EC 4.1.2.53),35,36 to generate the S-3-
substituted 4-hydroxy-2-oxocarboxylic acid (S-3) precursors, and 3-methyl-2-
oxobutanoate hydroxymethyltransferase (KPHMT, EC 2.1.2.11), to furnish the
ACS Paragon Plus Environment
6

Page 7 of 37
ACS Catalysis
1
2
3
4
5
6
7
8
9
corresponding enantiomers (R-3). In a second step, 3 are transformed in situ into 2-
substituted 3-hydroxycarboxylate sodium salt by oxidative decarboxylation (Scheme 1).
Finally, esterification with 2,4′-dibromoacetophenone, benzyl or 4-nitrobenzyl bromide,
makes them useful and manageable chiral building blocks and, in turn, facilitates the
isolation and purification of the products as 2-(4-bromophenyl)-2-oxoethyl (4), benzyl (5)
or 4-nitrobenzyl (6) esters. 2-(4-Bromophenyl)-2-oxoethyl ester is a convenient protecting
group because of the mild conditions used for its introduction and removal (e.g., by
treatment of the derivative with zinc in acetic acid or using visible light photoredox
catalysis).37-41 Furthermore, most of 2-(4-bromophenyl)-2-oxoethyl esters are solids, easy
to crystallize and as such, they can be characterized by crystallography. Moreover, more
synthetically convenient benzyl and 4-nitrobenzyl esters can also be obtained without
compromising the enantiomeric purity of the products, since the esterification was
conducted in absence of base.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment
7

ACS Catalysis
Page 8 of 37
1
2
3
O
4
5
6
7
8
9
CO2Na
OH
O
O
OH
R1
H
1
+
O
a) KPHMT
a) MBP-YfaU
CO2Na
NaO2C
R1:
:-CH3
-CH CH
-CH(CH3)2
-(CH2)2CH3
O
R1:
R1
R1
3
R-
3
S-
a
e
f
g
h
:
-(CH2)3CH3
:-CH2CH=CH2
:-(CH2)2CH=CH2
:-(CH2)3CH=CH2
H
2
b
c
d
:
:
:
2
3
b) H2O2, -CO2-
c) R2Br
b) H2O2, -CO2-
c) R2Br
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
i
j
k
:
:
:
O
OH
OH
R1
O
R2O
OR2
R2
=
R1
4
R-
4
S-
Br
Scheme
1.
Stereodivergent
chemoenzymatic
synthesis
of
2-substituted
3-hydroxycarboxylate esters (4). a) Aldol addition of 2-oxoacids to methanal; b)
decarboxylation
with
H2O2,
c)
ester
formation
by
esterification
with
2,4′-dibromoacetophenone.
Results and discussion
Recently, we reported wild-type Class II YfaU aldolase from E. coli, fused with the
maltose binding protein (MBP) from E. coli (i.e., MBP-YfaU), and variants thereof as
robust biocatalysts for aldol additions of 2-oxoacids to a variety of electrophiles.35,36
Moreover, MBP-YfaU is unique among aldolases in that it retains activity at methanal
concentration >1 M and, therefore, a highly qualified biocatalyst for the planned
reactions.42 Moreover, KPHMT is a metal cofactor dependent enzyme that utilizes
5,10-methylene-5,6,7,8-tetrahydrofolate to transfer a hydroxymethyl group to 3-methyl-2-
ACS Paragon Plus Environment
8

Page 9 of 37
ACS Catalysis
1
2
3
4
5
6
7
8
9
oxobutanoate to furnish ketopantoate (i.e., 4-hydroxy-3,3-dimethyl-2-oxobutanoic
acid).43,44 Interestingly, KPHMT can function as a Class II aldolase catalyzing the aldol
addition of 3-methyl-2-oxobutanoate to methanal operating under a tetrahydrofolate
independent mechanism.45 This is a major change in the catalytic mechanism of this
enzyme, which was uncovered by following the rate of deuterium exchange of the β-
protons during 2-oxoacid enolization.45 A similar dual mechanism is found in pyridoxal
5’-phosphate dependent serine hydroxymethyl transferases (SHMT, E.C. 2.1.2.1).46 We
report herein the first synthetic application of the promiscuous activity of KPHMT as an
aldolase type biocatalyst. It is also remarkable the fact that methanal, that is normally
toxic to proteins and deactivates enzymes, even at low concentrations, can be utilized as
electrophile with these enzymes in an enantioselective fashion.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Different methodologies are reported in the literature on synthesis of 2-oxoacids.47,48 In
this work, the preparation of the selected 2-oxoacids 1 was accomplished using the
Grignard reaction49 starting with the corresponding bromoalkane, formation of the alkyl-
magnesium bromide, addition to diethyloxalate and finally, ethyl ester hydrolysis
catalyzed by Novozym 435 lipase. Combination of these chemical and enzymatic
ACS Paragon Plus Environment
9

ACS Catalysis
Page 10 of 37
1
2
3
4
5
6
7
8
9
procedures resulted in a simple way to obtain 1d-h and 1k in 67-75% isolated yields. The
synthesis of 1i-j, was achieved by alkylation of ethyl 1,3-dithiane-2-carboxylate with the
corresponding bromocycloalkane, followed by oxidative deprotection of the carbonyl
group using N-bromosuccinimide and finally, ethyl ester hydrolysis catalyzed by Novozym
435 lipase. This procedure rendered the 2-oxoacids in 39 and 70% isolated yields,
respectively (see SI pages S15-S47).50,51
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
We initiated the study with the screening of the aldol addition of the 2-oxoacids 1 to 2
using wild-type MBP-YfaU(Ni2+) and KPHMT(Co2+) as catalysts, with the metal cofactors
that gave the best activity (see SI page S14).42
Wild-type MBP-YfaU displayed a significant activity (v0) for 1a but considerably lower
for 1b and 1i, and much lower for the other nucleophiles (Figure 3A). This resulted in good
product formation at 24 hours for 1a-b and 1i, but low conversions for 2-oxoacid
nucleophiles with C3-substitutions larger than ethyl or cyclopropyl (Figure 3C). Similar
observations had already been reported, which were attributed to size limitations of the
MBP-YfaU nucleophile binding site.35 Indeed, a molecular model of the ternary complex
between wild-type YfaU, the 1i-enolate and 2 shows a good accommodation of the
ACS Paragon Plus Environment
10

Page 11 of 37
ACS Catalysis
1
2
3
4
5
6
7
8
9
substrates in the active site (Figure 4A). However, fitting larger nucleophiles would imply
conformations that are more strained and/or the appearance of clashes with active site
residues. MBP-YfaU W23V and the double mutant MBP-YfaU W23V/L216A, designed to
increase the volume of the nucleophile binding site,35 showed improved product
conversions, i.e. 74–97% (Figure 3A). MBP-YfaU W23V consistently showed moderately
higher conversions than the double mutant towards larger substituents, e.g. 1d-e, 1h and
1k, as well as higher reaction rates (Figure 3A and 3C). As shown in Figure 4B, the single
W23V mutant can nicely accommodate even the largest cyclic substitution assayed (1k)
in the cavity generated by the mutation, without perturbing the approach of the methanal
molecule.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment
11

ACS Catalysis
Page 12 of 37
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 3. Initial reaction rates determined for the aldol reaction of 2-oxoacids (1) to
methanal catalyzed by MBP-YfaU (Panel A) and KPHMT (Panel B). 1Only determined for
the wild-type enzyme. 2Not determined for the I202A variant. Aldol product formation (%,
measured by HPLC) at 24 h of the aldol addition reaction of 2-oxoacids (1) to methanal
ACS Paragon Plus Environment
12

Page 13 of 37
ACS Catalysis
1
2
3
4
5
6
7
8
9
(2) catalyzed by MBP-YfaU (Panel C) and KPHMT (Panel D). Conditions: MBP-YfaU
catalysts (4 mg protein mL–1 in the reaction, 0.06 mol%)35,36 plus NiCl2 (0.25 mM in the
reaction, 4 eq with respect to the enzyme), or KPHMT catalysts (glycerinated protein
solution, 4 mg protein mL–1 in the reaction, 0.14 mol%) plus CoCl2 (0.6 mM in the reaction,
4 eq with respect to the enzyme), and 2-oxoacid (1a-k, 0.05 mmol, 1 eq, 0.1 M in the
reaction). No specific buffer was employed, the 2-oxoacid adjusted at pH 7.0 buffered the
solution. The reactions were started by adding methanal (2, 3.8 µL of a 13.3 M commercial
solution, 0.05 mmol, 1 eq, 0.1 M in the reaction). The total reaction volume was 0.5 mL.
Error bars are the values of the standard error of the mean of two independent
experiments under the same reaction conditions. aNot determined.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment
13

ACS Catalysis
Page 14 of 37
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 4. Models of (A) wild-type YfaU, (B) YfaU W23V, (C) wild-type KPHMT, (D) KPHMT
I212A, (E) KPHMT I202A and (F) KPHMT L42A/I212A complexed with formaldehyde (2,
ACS Paragon Plus Environment
14

Page 15 of 37
ACS Catalysis
1
2
3
4
5
6
7
8
9
green C-atoms) and nucleophiles (yellow C-atoms) (A,C) 1i, (B) 1k, (D,F) 1h and (E) 1c.
Mutated residues are highlighted with orange C-atoms. Nucleophiles are shown in their
enolate form coordinated to the metal cofactor (purple sphere), which is also coordinated
by two water molecules and by residues E153 and D179 of YfaU or D45 and D84 of
KPHMT. Methanal is disposed with its carbonyl O-atom forming hydrogen bonds with
residue R74 in YfaU, or with residue E181 in KPHMT, as well as with one of the metal-
coordinated waters. Surfaces of the active site cavities near to the nucleophile substrates
are shown in cyan. Interactions are shown with dashed lines: H-bond in yellow, salt
bridges in magenta. Models were built as described in the Supporting Information, starting
from the crystal structure of YfaU in complex with pyruvate (PDB 2VWT)44,52 and of
KPHMT in complex with ketopantoate (PDB 1M3U).44
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Wild-type KPHMT showed substantial activities (v0) and good conversions (80-97%) for
nucleophiles with substituents such as methyl (1a), ethyl (1b), allyl (1f) or cyclopropyl (1i),
whereas it exhibited low efficiency with branched (1c), long alkyl or alkenyl (1d-e, g-h) or
larger alicyclic (C>3) substituents (1j-k) (0-24%) (Figure 3B and 3D). As already shown
ACS Paragon Plus Environment
15

ACS Catalysis
Page 16 of 37
1
2
3
4
5
6
7
8
9
for MBP-YfaU, a molecular model of the complex between wild-type KPHMT, the 1i-
enolate and 2 shows a relatively tight packing (i.e., distance < 4 Å) of the nucleophile
substituent R1 (i.e., cyclopropyl) against a hydrophobic cavity delimited by the side chains
of residues L42, H136, V179, E181, I202 and I212 (Figure 4C). To increase the size of
this cavity, residues I202 and I212 were initially selected for simple conservative
replacement by alanine. KPHMT I212A variant accepted nucleophiles with long chain
substituents (1d-h), showing highly improved reaction rates relative to wild-type and
furnishing high conversions (88-94%) (Figure 3B and 3D), but it was not competent with
branched (1c) or larger cyclic (1j-k) derivatives. Fortunately, the I202A mutation displayed
some activity with these later types of nucleophiles, which resulted on reasonable
conversions for 1c (57%) and 1j (79%), while keeping the activity for smaller linear or
cyclic substrates, such as 1e or 1i (Figure 3B and 3D). Modeling the effects of these
mutations on the size and shape of the active site suggests that the I212A mutation
generates a deeper cavity at the nucleophile site, which can adequately fit longer
hydrophobic chains like the one present in 1h (Figure 4D). Furthermore, the I202A
mutation extends and increases the volume of the active site towards one side, allowing
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment
16

Page 17 of 37
ACS Catalysis
1
2
3
4
5
6
7
8
9
an adequate placement of branched substrates such as 1c (Figure 4E). Nevertheless, the
I202A mutant was still inefficient with the largest cyclic substitution (1k). Further attempts
to improve the activity of KPHMT implied the L42A mutation alone or combined with the
I212A one. However, none of these two variants revealed any activity with 1k and the
single mutant (L42A) even displayed a deleterious effect relative to the wild-type enzyme
(Figure 3D). In this regard, it cannot be discarded that differences in the conversions
between enzyme variants could also arise from inhibitory effects of substrates and
products.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The best biocatalyst for each case was selected to run preparative scale experiments
and evaluate the isolated yields, enantiomeric excesses, and absolute stereochemistry
(Table 1). Enantiomeric excess (ee) of R- and S-4a-k were determined by chiral HPLC
against racemic samples obtained by chemical synthesis after derivatization with
2,4’-dibromoacetophenone and purification (see SI Figures S128-S141). The substrate
concentration of the reactions was adjusted to achieve the maximum productivity and
enantiomeric excess (see SI pages S48-S91). Using KPHMT, we observed for 1a and 1b
that on increasing the substrate concentration, the ee decreased due to the enzyme
ACS Paragon Plus Environment
17

ACS Catalysis
Page 18 of 37
1
2
3
4
5
6
7
8
9
inactivation by methanal, and a major contribution of a non-enantioselective background
reaction catalyzed by the free metal cofactor (see SI Figure S64 page 73; and Figure S68
page S75). For the rest, the influence of the background reaction was negligible. The
absolute stereochemistry was determined using the following procedures: i) by the
synthesis of authentic samples of R- and S-4a from commercially available (R)- and (S)-
β-hydroxyisobutyric acid sodium salts (see SI Figure S128), ii) by comparing the specific
rotation values of the compounds with those reported in the literature (R and S-4b-g) (see
SI Table S7) and iii) by X-ray diffraction for those uncovered compounds (i.e. compounds
4i-k) (see SI pages S154-S160). The absolute configuration of products S-4h and R-4h
was inferred based on the high stereopreference observed for the enzymes MBP-YfaU
and KPHMT found in the examples thereof.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Table 1. Preparative scale synthesis of 2-(4-bromophenyl)-2-oxoethyl 2-substituted-3-
hydroxycarboxylate esters (4) after MBP-YfaU and KPHMT-catalyzed aldol addition of
2-oxoacids (1a-k) to methanal (2), decarboxylation and esterification of 3.
ACS Paragon Plus Environment
18

Page 19 of 37
ACS Catalysis
1
2
3
4
O
O
5
6
7
8
CO2Na
OH
O
O
OH
R1
H
1
+
O
a) KPHMT
a) MBP-YfaU
R2 =
OR2
R2O
b) H2O2, -CO2-
c) R2Br
b) H2O2, -CO2-
c) R2Br
R1
Br
R1
4
4
S-
R-
9
H
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2
[2] = [1]
/M
2-Oxoacid
Isolated ee/%[
Biocatalyst
Product (4)
a]
(1)
Yield/%
1a
1a
MBP-YfaUwt
KPHMTwt
1.0
0.2
0.4
1.0
0.2
1.0
0.1
1.0
0.1
0.6
0.1
0.1
1.0
0.2
S-4a 80
98
98
95
99
95
99
92
99
95
99
86
93
98
96
CO2R2
HO
76
R-4a
88
MBP-YfaUW23V
KPHMTwt
S-4b 81
R-4b 63
S-4c 75
R-4c 49
S-4d 68
R-4d 60
S-4e 65
R-4e 55
R-4e 57
S-4f 73
R-4f 68
CO2R2
CO2R2
CO2R2
1b
1b
1c
1c
1d
1d
1e
1e
1e
1f
HO
HO
HO
MBP-YfaUW23V
KPHMTI202A
MBP-YfaUW23V
KPHMTI212A
MBP-YfaUW23V
KPHMTI212A
CO2R2
HO
HO
KPHMTL42A/I212A
MBP-YfaUW23V
KPHMTI212A
CO2R2
1f
ACS Paragon Plus Environment
19

ACS Catalysis
Page 20 of 37
1
2
3
4
5
6
7
8
9
1g
1g
1h
1h
1h
1i
MBP-YfaUW23V
KPHMTI212A
0.8
0.2
0.6
0.2
0.2
1.0
0.6
0.4
0.8
0.2
S-4g 69
R-4g 68
S-4h 68
R-4h 72
R-4h 64
S-4i 73
R-4i 63
R-4i 68
S-4j 80
R-4j 48
95
97
78
88
93
98
83
87
94
75
CO2R2
HO
MBP-YfaUW23V
KPHMTI212A
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
CO2R2
HO
KPHMTL42A/I212A
MBP-YfaUW23V
KPHMTI202A
CO2R2
HO
1i
1i
KPHMTwt
CO2R2
CO2R2
1j
MBP-YfaUW23V
KPHMTI202A
HO
HO
1j
1k
MBP-YfaUW23V
0.3
S-4k 73
96
[a]Determined by HPLC on a chiral stationary phase. The progress curves for these
preparative scale reactions and the HPLC monitoring on reverse phase of the reactions
are given in Figures S18-S101 in the Supporting Information.
MBP-YfaU(Ni2+) gave very good enantioselectivities (94-96%) for 1g, 1j and 1k and
excellent (98-99%) for 1a-f and 1i at 0.6-1.0 M equimolar substrate concentration. It was
also observed that on increasing the length of the alkenyl substituent from butenyl (1g) to
ACS Paragon Plus Environment
20

Page 21 of 37
ACS Catalysis
1
2
3
4
5
6
7
8
9
pentenyl (1h) the ee decreased from 95 to 78%. Since the background reaction was
negligible in these cases and given the requirements of the enzymatic mechanism (i.e.,
metal-coordinated nucleophile enolate and methanal accepting a hydrogen bond from the
metal-coordinated water), this was likely due to an increased occurrence of the inverted
orientation of the R1 substituent (i.e., giving rise to a Z-enolate) owing to the steric
interactions with the protein. Alternatively, the possibility that the nucleophile is
accommodated in an inverted orientation, with the side chain R1 pointing towards the
active site entrance, and exposing the pro-R enantiotopic face to the electrophile, cannot
be excluded.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The best KPHMT catalysts exhibited very good ee values (92-98%) for nucleophiles 1a-
h. An equimolar substrate concentration at 0.1 or 0.2 M was best to achieve the maximum
product formation under the selected reaction conditions (Table 1). For 1a-b, 1d, 1f and
1g, wild-type KPHMT and the I212A variant gave ee values between 95-98%. For 1e and
1h the ees were better for L42A/I212A than those with I212A. For branched 1c and cyclic
1j nucleophile substituents, the I202A variant gave 92% and 75% ee, respectively.
Cyclopropyl (1i) was a particular case since it was tolerated for the wild-type and I202A,
ACS Paragon Plus Environment
21

ACS Catalysis
Page 22 of 37
1
2
3
4
5
6
7
8
but both with moderate ee (83-87%) as compared with the rest of the selected
nucleophiles.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Besides 2-(4-bromophenyl)-2-oxoethyl esters 4, benzyl 5 and 4-nitrobenzyl esters 6
(Scheme 2) can also be produced using the corresponding BnBr/4-NO2C6H4CH2Br in the
presence of CsCl without compromising the enantiomeric purity.
O
CO2Na
OH
O
O
OH
R1
H
1
+
O
a) KPHMT
a) MBP-YfaU
R2 =
OR2
R2O
b) H2O2, -CO2-
c) R2Br, CsCl
b) H2O2, -CO2-
c) R2Br, CsCl
R1
S-
R1
NO2
5a
6a
5g
R-
R-
R-
5a
6a
S-
H
2
O
O
O
O
O
HO
O
HO
O
HO
OBn
HO
OBn
HO
OBn
NO2
NO2
5a
6a
5a
6a
S-
S-
R-
R-
R-
5g
(60%, 99% ee)
(60%, 98% ee)
(40%, 99% ee)
(59%, 98% ee)
(60%,
97%
ee)
Scheme 2. Synthesis of benzyl and 4-nitrobenzyl esters 5 and 6 obtained from R-3 and
S-3 aldol adduct precursors. For experimental details and product characterization data
see pages S126 to S135 in Supporting Information.
ACS Paragon Plus Environment
22

Page 23 of 37
ACS Catalysis
1
2
3
4
5
6
7
8
9
In summary, a convenient and versatile three-step chemoenzymatic method for the
synthesis of enantiomerically pure 2-substituted 3-hydroxycarboxylate esters (4, 5 and 6)
was developed. This novel chemo enzymatic methodology provides high isolated yields
and enantiomeric excesses through simple esterification reaction and work up
procedures. The key step in the synthetic pathway is the aldol addition of 2-oxoacids (1)
to methanal (2) catalyzed by two enantiocomplementary 2-oxoacid carboligases; MBP-
YfaU(Ni2+) and, for the first time, KPHMT(Co2+) operating under a tetrahydrofolate
independent mechanism, to furnish (S)- and (R)-3-substituted 4-hydroxy-2-oxoacid
precursors, respectively. The nucleophile tolerance towards a broad range of structurally
diverse 2-oxoacids of wild-type MBP-YfaU(Ni2+) and KPHMT(Co2+) was tuned by using
structurally-guided site directed mutagenesis: for MBP-YfaU(Ni2+) a single W23V
mutation was sufficient to accommodate even the largest cyclohexane substitution (1k),
without perturbing the approach of the electrophile. For KPHMT(Co2+) three variants,
I202A, I212A and L42A/I212A, were constructed to accomplish the required nucleophile
tolerance. In this case, mutation I212A was critical to generate a deeper cavity at the
nucleophile site, which can adequately fit longer hydrophobic chains, although insufficient
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment
23

ACS Catalysis
Page 24 of 37
1
2
3
4
5
6
7
8
9
with the largest cyclic substitution (1k). For both aldolases, excellent (94-99% ee) to very
good (88-98% ee) stereoselectivities and 57–88% isolated yields were achieved at
substrate concentrations in a range between 0.1–1.0 M. It is noteworthy that the variants
of both enzymes follow the same stereochemical outcome as that of the wild-type, and
therefore the mutations did not affect their stereo preference. The limitations encountered
in other methods reported, such as harsh conditions, e.g. high H2 and CO pressures, poor
structural versatility (i.e. some methods limited to Roche ester, e.g. Baylis-Hillman plus
ene reduction sequence) and high organocatalyst loading (e.g. 30 mol%) can be
overcome by this methodology. The broad nucleophile substrate tolerance exhibited by
both 2-oxoacid utilizing carboligases allow to envisage further synthetic possibilities of
these highly enantioselective enzymes. For instance, the intermediate precursors of 2-
substituted 3-hydroxycarboxylate esters are 4-hydroxy-2-keto acid derivatives, which are
versatile building blocks for the synthesis of amino acids, hydroxyl carboxylic acids and
aldehydes among others overcoming side reactions such as polymerization, homo
aldolizations and aldol condensations, as well as low stereoselectivities, usually
encountered with conventional chemical or organocatalytic methods.53-56 In addition, from
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment
24

Page 25 of 37
ACS Catalysis
1
2
3
4
5
6
7
8
9
an organic chemistry process point of view, it would be interesting to consider the option
of running the reactions using a continuous addition of methanal since it has been
demonstrate to be highly beneficial in previous reports.42,57
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Experimental
Typical procedure: Synthesis of 2-(4-bromophenyl)-2-oxoethyl (S)-3-hydroxy-2-
methylpropanoate (S-4a) Reaction (2 mL total volume) was carried out at room
temperature with magnetic stirring (250 rpm). The reactions were conducted in a round-
bottom flask of 100 mL capacity to avoid spilling of the reaction media due to the huge
amount of foam generated when adding the catalase to eliminate the excess of H2O2 (see
below). MBP-YfaU wild-type (16 mg, 4 mg protein mL–1 in the reaction, 0.005 mol%) was
dissolved in a solution of sodium 2-oxobutanoate (1a, 1 mL of a 2 M stock solution pH
7.0, 2 mmol, 1 eq, 1 M in the reaction). No specific buffer was employed, 1a adjusted at
pH 7.0 acted as buffer. Plain water (0.85 mL) and NiCl2 (5 µL of a 0.1 M stock solution,
0.25 mM in the reaction) were added. The reaction was started by adding methanal (2,
ACS Paragon Plus Environment
25

ACS Catalysis
Page 26 of 37
1
2
3
4
5
6
7
8
9
0.15 mL of a 13.3 M commercial solution, 2 mmol, 1 eq, 1 M in the reaction). When the
reaction was finished hydrogen peroxide (1.14 mL of a 8.8 M commercial solution, 10
mmol, 5 eq) was added and the reaction led to proceed until no aldol adduct was detected
by HPLC. Then, catalase from bovine liver (50 mg) dissolved in 1 mL of 10 mM sodium
phosphate buffer pH 7.0 was added to eliminate the excess of H2O2. The reaction mixture
was diluted with methanol (20 mL), filtered through Celite® and the pellet was washed
with methanol (3x50 mL). Glycerol (2-3 mL) was added to solubilize the carboxylic sodium
salt residue once the solvent was removed under vacuum. Then, the mixture was diluted
with DMF (10 mL) and 2,4’-dibromoacetophenone (556 mg, 2 mmol, 1 eq) was added.
When 2,4’-dibromoacetophenone was consumed (monitored by HPLC), EtOAc (200 mL)
was added to the reaction mixture and washed with H2O (3x50 mL). The combined
organic phases were dried over anhydrous MgSO4, absorbed onto silica gel (50 g) and
loaded onto a flash chromatography on silica. Product was eluted with a step gradient of
hexane:EtOAc. Pure fractions were pooled and the solvent removed under vacuum
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
affording S-4a as a white solid (482 mg, 80%) [α]20 = + 19.0 (c = 1 in MeOH). See
D
Supporting Information page S92 for product characterization.
ACS Paragon Plus Environment
26

Page 27 of 37
ACS Catalysis
1
2
3
4
5
6
7
8
9
Typical procedure (R-4a): Reaction (2 mL total volume) was carried out at room
temperature with magnetic stirring (250 rpm). The reactions were conducted in a round-
bottom flask of 100 mL capacity to avoid spilling of the reaction media due to the huge
amount of foam generated when adding the catalase to eliminate the excess of H2O2. To
a solution of KPHMT wild-type (0.75 mL of a 10 mg mL–1 stock glycerinated protein
solution, 4 mg mL–1 in the reaction, 0.03 mol%), sodium 2-oxobutanoate (1a, 0.8 mL of a
1 M stock solution pH 7.0, 0.8 mmol, 1 eq, 0.4 M in the reaction) was added. No specific
buffer was employed, 1a adjusted at pH 7.0 acted as buffer. Plain water (0.37 mL) and
CoCl2 (12 µL of a 0.1 M stock solution, 0.6 mM in the reaction) were added. A
homogeneous white colloidal solution appeared. The reaction was started by adding
methanal (2, 60 µL of a 13.3 M commercial solution, 0.8 mmol, 1 eq, 0.4 M in the reaction).
From that point, the procedure was identical to that the one described for the S-
enantiomers described above. The product R-4a was obtained as a white solid (214 mg,
88%). [α]20D = – 18.5 (c = 1 in MeOH). See Supporting Information page S93 for product
characterization.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment
27

ACS Catalysis
Page 28 of 37
1
2
3
4
5
ASSOCIATED CONTENT
6
7
8
9
AUTHOR INFORMATION
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Corresponding Author
*E-mail for PC: pere.clapes@iqac.csic.es.
Author Contributions
The manuscript was written through contributions of all authors. All authors have given
approval to the final version of the manuscript.
Funding Sources
European Union’s Horizon 2020 research and innovation program under grant
agreement No 635595 (CarbaZymes), Ministerio de Economía y Competitividad
(MINECO), Fondo Europeo de Desarrollo Regional (FEDER) (grant RTI2018-094637-B-
I00), and Programación Conjunta Internacional PCI2018-092937, through the initiative
ERA CoBioTech.
ACS Paragon Plus Environment
28

Page 29 of 37
ACS Catalysis
1
2
3
4
5
6
7
8
9
Supporting Information. Materials, general procedures, biocatalysts YfaU and KPHMT
wild-type and variants expression and purification, activity assay of KPHMT catalysts,
preparation of 2-oxoacids, racemic aldol adducts, biocatalysts screening, preparative
biocatalytic reactions, full preparative reactions, HPLC and chiral HPLC analysis,
description of physical and spectral properties and NMR spectra of the final compounds,
computational methods (PDF). This information is available free of charge on the ACS
Publications website.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACKNOWLEDGMENT
This project has received funding from the European Union’s Horizon 2020 research
and innovation program under grant agreement No 635595 (CarbaZymes), the
Ministerio de Economía y Competitividad (MINECO), the Fondo Europeo de Desarrollo
Regional (FEDER) (grant RTI2018-094637-B-I00), Programación Conjunta
Internacional PCI2018-092937, through the initiative ERA CoBioTech. The authors
gratefully acknowledge the “Consorci de Serveis Universitaris de Catalunya” (CSUC) for
allowing the use of its software and hardware resources.
ACS Paragon Plus Environment
29